Fmi-weather-client

What's Changed
* Fix parsing of xml data by fursov in https://github.com/saaste/fmi-weather-client/pull/59
* Update `xmltodict` to 0.14.1 by fursov in https://github.com/saaste/fmi-weather-client/pull/59
* Drop Python 3.8 support by saaste in https://github.com/saaste/fmi-weather-client/pull/62
* Drop `requests` to 2.32.3 by saaste in https://github.com/saaste/fmi-weather-client/pull/62

New Contributors
* fursov made their first contribution in https://github.com/saaste/fmi-weather-client/pull/59

**Full Changelog**: https://github.com/saaste/fmi-weather-client/compare/0.3.0...0.4.0

What's Changed
- 56: Upgrade requests (fixes [CVE-2024-35195](https://github.com/advisories/GHSA-9wx4-h78v-vm56))
- **Drop Python 3.7 support**

**Full Changelog**: https://github.com/saaste/fmi-weather-client/compare/0.2.2...0.3.0

What's Changed
* 51: Fix invalid coordinate handling by saaste in https://github.com/saaste/fmi-weather-client/pull/54 (reported by kallere)


**Full Changelog**: https://github.com/saaste/fmi-weather-client/compare/0.2.1...0.2.2

What's Changed
* Added FMI "feels like" temperature calculation. By jaenis in https://github.com/saaste/fmi-weather-client/pull/52

New Contributors
* jaenis made their first contribution in https://github.com/saaste/fmi-weather-client/pull/52 🥳

**Full Changelog**: https://github.com/saaste/fmi-weather-client/compare/0.2.0...0.2.1

Support for the old Hirlam weather model is going to end in September 2022. Fixes 39.

From FMI website (https://en.ilmatieteenlaitos.fi/hirlam-opendata-on-aws-s3):

> The use of the HIRLAM weather forecasting model at the Finnish Meteorological Institute will be discontinued in September 2022, after which its forecasts are no longer available through the open data services.
>
> During the past fifteen years, the main focus in forecast model development has been on a more accurate model with improved resolution. The resulting HARMONIE (MEPS) model is capable of 2.5 km horizontal resolution (compared to 7.5 km of HIRLAM) and able to predict small scale phenomena more accurately. HARMONIE (MEPS) forecasts are now available through the open data services.
>
> More information will be made available later, but already now it is worth switching to use HARMONIE (MEPS) model if possible.

The following data is not provided by Harmonie so values will always be None:
- wind_max
- radiation_long_wave_acc
- radiation_short_wave_diff_surface_acc
- land_sea_mask

Fields are kept in the model for backward compatibility.